Background: Fibrates are lipid-lowering agents that act as peroxisome proliferator-activated receptor (PPAR)-α agonists. They have been associated with cancers in experimental models, but data in humans are rare, and among published studies none has investigated cancers in tissues with high PPAR-α concentrations.
Methods: A nested case-control study was performed in a French population-based healthcare database. Adults aged ≥45 years, and free of cancer for 3 years, were followed for 5 years for incident cases of melanoma, non-melanoma skin cancers, thyroid, pancreas, bladder, and kidney cancers. Cases were matched with up to ten controls for age, sex, and diseases that could increase the risk of cancers. Conditional logistic models, adjusted for drug-use as potential confounders, were used to estimate the risk (odds ratio [OR]) of cancers of interest (and individual cancers) associated with cumulative exposure to fibrates (defined daily doses [DDD]).
Results: Among the 147,338 eligible subjects, 3,331 (2.3 %) cases of studied cancers were identified. Only use of fibrates >550 DDDs was associated with an increased risk (OR 1.26; 95 % CI 1.12-1.42), and similar results were found for statins (≥1,460 DDDs; OR 1.15; 95 % CI 1.03-1.28). When considering cancers individually, the association was significant for non-melanoma skin-cancer (OR 1.35; 95 % CI 1.14-1.60), and close to significance for bladder cancer (OR 1.26; 95 % CI 0.96-1.64); similar associations with the use of statins were found.
Conclusions: The associations found between fibrate exposure and cancers of tissues with high PPAR-α concentrations were most likely related to residual confounding as similar associations were found for statins.